feat(gopt): add remove redundant copy pass

GitOrigin-RevId: f616a76d29

src/gopt/impl/framework.cpp

@@ -649,6 +649,7 @@ GraphOptimizer& GraphOptimizer::add_preset_passes(
     add_pass<ReorderArithChainPass>(cv_type);
     add_pass<FinalArithTransformPass>();
     add_pass<RemoveRedundantTypeCvtPass>();
+    add_pass<RemoveRedundantCopyPass>();
 
 #if MGB_JIT
     bool need_jit = false;

src/gopt/impl/misc.cpp

@@ -682,6 +682,69 @@ void RemoveRedundantTypeCvtPass::apply(OptState& opt) const {
     MIDOUT_E
 }
 
+/* ======================= RemoveRedundantCopyPass ====================== */
+
+const char* RemoveRedundantCopyPass::name() const {
+    return "remove_redundant_copy";
+}
+
+bool RemoveRedundantCopyPass::should_remove(const CompNode& A,
+                                            const CompNode& B) {
+    //! if A and B has the same memnode and cpu <-> atlas/cpu <-> cuda, as only
+    //! these two compnode support crosscncopy
+    if (A.mem_node() == B.mem_node() ||
+        ((A.device_type() == CompNode::DeviceType::CPU ||
+          A.device_type() == CompNode::DeviceType::MULTITHREAD) &&
+         (B.device_type() == CompNode::DeviceType::ATLAS ||
+          B.device_type() == CompNode::DeviceType::CUDA)) ||
+        ((B.device_type() == CompNode::DeviceType::CPU ||
+          B.device_type() == CompNode::DeviceType::MULTITHREAD) &&
+         (A.device_type() == CompNode::DeviceType::ATLAS ||
+          A.device_type() == CompNode::DeviceType::CUDA))) {
+        return true;
+    } else {
+        return false;
+    }
+}
+
+void RemoveRedundantCopyPass::apply(OptState& opt) const {
+    MIDOUT_B("RemoveRedundantCopyPass::apply")
+    auto rewriter = opt.graph().make_rewriter();
+
+    auto on_opr = [&](OperatorNodeBase* opr) {
+        if (auto copy0 = try_cast_as_op<opr::Copy>(opr)) {
+            auto inp0 = rewriter.get_var(copy0->input(0));
+            if (auto copy1= try_cast_as_op<opr::Copy>(inp0)) {
+                auto inp1 = copy1->input(0);
+                if (should_remove(inp1->comp_node(),
+                                  copy0->output(0)->comp_node())) {
+                    mgb_assert(!rewriter.has_manual_replace(inp1));
+                    if (inp1->comp_node() == copy0->output(0)->comp_node()) {
+                        rewriter.replace_var(
+                                copy0->output(0), inp1,
+                                mgb_cstr_log("copy(copy(a0, a1), a0) -> "
+                                             "a0"));
+                        return;
+                    } else {
+                        auto fold = opr::Copy::make(
+                                inp1, copy0->output(0)->comp_node());
+                        rewriter.replace_var(
+                                copy0->output(0), fold.node(),
+                                mgb_cstr_log("copy(copy(a0, a1), a2) -> "
+                                             "copy(a0, a2)"));
+                        return;
+                    }
+                }
+            }
+        }
+        rewriter.auto_replace_outputs(opr);
+    };
+
+    opt.graph().iter(on_opr);
+    rewriter.apply_inplace();
+    MIDOUT_E
+}
+
 #if MGB_ENABLE_OPR_MM
 #include "megbrain/opr/collective_comm.h"
 

src/gopt/include/megbrain/gopt/misc.h

@@ -85,6 +85,16 @@ namespace gopt {
         void apply(OptState &opt) const override;
     };
 
+    class RemoveRedundantCopyPass final : public Pass {
+    private:
+        //! Remove the copy chain of form cpu -> cpu -> cpu,
+        //! cpu -> gpu -> cpu
+        static bool should_remove(const CompNode& A, const CompNode& B);
+    public:
+        const char * name() const override;
+        void apply(OptState &opt) const override;
+    };
+
     //! remove execution mask for const PPVs in conditional execution
     class CondExecConstPredicateFolding final : public Pass {
     public:

src/gopt/test/helper.h

@@ -26,14 +26,16 @@ namespace mgb {
             HostTensorGenerator<> gen;
             std::shared_ptr<ComputingGraph> graph = ComputingGraph::make();
 
-            SymbolVar mkvar(const char *name, const TensorShape &shp = {1}) {
-                return opr::Host2DeviceCopy::make(
-                        *graph, gen(shp)).rename(name);
+            SymbolVar mkvar(const char* name, const TensorShape& shp = {1},
+                            CompNode cn = CompNode::load("xpu0")) {
+                return opr::Host2DeviceCopy::make(*graph, gen(shp), cn)
+                        .rename(name);
             }
 
-            SymbolVar mkcvar(const char *name, const TensorShape &shp = {1}) {
+            SymbolVar mkcvar(const char* name, const TensorShape& shp = {1},
+                             CompNode cn = CompNode::load("xpu0")) {
                 return opr::SharedDeviceTensor::make(
-                        *graph, *gen(shp)).rename(name);
+                        *graph, *gen(shp), cn).rename(name);
             }
 
             template<typename ...Args>
@@ -73,4 +75,3 @@ namespace mgb {
     TEST_F(TestGopt##pass, name)
 
 // vim: syntax=cpp.doxygen foldmethod=marker foldmarker=f{{{,f}}}
-

src/gopt/test/misc.cpp

@@ -16,6 +16,7 @@
 #include "megbrain/opr/basic_arith_wrapper.h"
 #include "megbrain/opr/blas.h"
 #include "megbrain/opr/cond.h"
+#include "megbrain/opr/io.h"
 #include "megbrain/opr/tensor_manip.h"
 #include "megbrain/opr/utility.h"
 
@@ -411,6 +412,71 @@ TEST_PASS(RemoveRedundantTypeCvtPass, Basic) {
     check(x_q8_q8, x_q8_fp32_q8_);
 }
 
+TEST_PASS(RemoveRedundantCopyPass, Basic) {
+    auto x = mkvar("x", {2, 3, 3}, CompNode::load("cpu0"));
+    {
+        auto x_cpu1 = opr::Copy::make(x, CompNode::load("cpu1"));
+        auto x_cpu0 = opr::Copy::make(x_cpu1, CompNode::load("cpu0"));
+        auto x_cpu2 = opr::Copy::make(x_cpu0, CompNode::load("cpu2"));
+        auto x_expected = opr::Copy::make(x, CompNode::load("cpu2"));
+        check(x, x_cpu0);
+        check(x_expected, x_cpu2);
+    }
+
+    {
+        auto x_cpu1 = opr::Copy::make(x, CompNode::load("cpu1"));
+        auto x_cpu2 = opr::Copy::make(x_cpu1, CompNode::load("cpu2"));
+        auto x_cpu3 = opr::Copy::make(x_cpu2, CompNode::load("cpu3"));
+        auto x_expected = opr::Copy::make(x, CompNode::load("cpu3"));
+        check(x_expected, x_cpu3);
+    }
+
+    {
+        auto x_cpu1 = opr::Copy::make(x, CompNode::load("cpu0:1"));
+        auto x_cpu2 = opr::Copy::make(x_cpu1, CompNode::load("cpu0:2"));
+        auto x_cpu3 = opr::Copy::make(x_cpu2, CompNode::load("cpu0:3"));
+        auto x_expected = opr::Copy::make(x, CompNode::load("cpu0:3"));
+        check(x_expected, x_cpu3);
+    }
+
+    {
+        auto x_cpu1 = opr::Copy::make(x, CompNode::load("cpu0:1"));
+        auto x_mt = opr::Copy::make(x_cpu1, CompNode::load("multithread8:0"));
+        auto x_cpu3 = opr::Copy::make(x_mt, CompNode::load("cpu0:3"));
+        auto x_expected = opr::Copy::make(x, CompNode::load("cpu0:3"));
+        check(x_expected, x_cpu3);
+    }
+
+#if MGB_ATLAS
+    {
+        auto x_atlas0 = opr::Copy::make(x, CompNode::load("atlas0"));
+        auto x_cpu2 = opr::Copy::make(x_atlas0, CompNode::load("cpu0:2"));
+        auto x_cpu3 = opr::Copy::make(x_cpu2, CompNode::load("cpu0:3"));
+        auto x_expected = opr::Copy::make(x, CompNode::load("cpu0:3"));
+        check(x_expected, x_cpu3);
+    }
+#endif
+
+#if MGB_CUDA
+    {
+        auto x_cuda0 = opr::Copy::make(x, CompNode::load("gpu0"));
+        auto x_cpu2 = opr::Copy::make(x_cuda0, CompNode::load("cpu0:2"));
+        auto x_cpu3 = opr::Copy::make(x_cpu2, CompNode::load("cpu0:3"));
+        auto x_expected = opr::Copy::make(x, CompNode::load("cpu0:3"));
+        check(x_expected, x_cpu3);
+    }
+
+    {
+        auto x_mt = opr::Copy::make(x, CompNode::load("multithread8:0"));
+        auto x_cpu2 = opr::Copy::make(x_mt , CompNode::load("gpu0:1"));
+        auto x_cpu3 = opr::Copy::make(x_cpu2, CompNode::load("multithread8:0"));
+        auto x_expected = opr::Copy::make(x, CompNode::load("multithread8:0"));
+        check(x_expected, x_cpu3);
+    }
+
+#endif
+}
+
 #if MGB_ENABLE_OPR_MM
 #include "megbrain/opr/collective_comm.h"
 #include "../../opr-mm/test/mock_client.h"